Main

In your final repo, there should be an R markdown file that organizes all computational steps for evaluating your proposed Facial Expression Recognition framework.

This file is currently a template for running evaluation experiments. You should update it according to your codes but following precisely the same structure.

Environment Setup

• Python env is required to run DNN model in R combined with Keras and Tensorflow
• Install Miniconda for Python (https://docs.conda.io/en/latest/miniconda.html)
• keras and tensorflow should be installed automatically
• use_condaenv(“r-tensorflow”) should be installed automatically. Restart RStudio if required in the console.

if(!require("EBImage")){
  install.packages("BiocManager")
  BiocManager::install("EBImage")
}
if(!require("R.matlab")){
  install.packages("R.matlab")
}
if(!require("readxl")){
  install.packages("readxl")
}

if(!require("dplyr")){
  install.packages("dplyr")
}
if(!require("readxl")){
  install.packages("readxl")
}

if(!require("ggplot2")){
  install.packages("ggplot2")
}

if(!require("caret")){
  install.packages("caret")
}

if(!require("glmnet")){
  install.packages("glmnet")
}

if(!require("WeightedROC")){
  install.packages("WeightedROC")
}

if(!require("gbm")){
  install.packages("gbm")
}

if(!require("xgboost")){
  install.packages("xgboost")
}

戼㸳̼愼㹤戼㸰昼㹣愼㸱愼㹥xgboost愼㸱愼㹦挼㹡挼㸷搼㸳挼㸳R戼㸰汾4.0.4 挼㸰戼㸴戼㹤愼㸸搼㸴攼㹣戼㸵挼㸴

if(!require("caret")){
  install.packages("caret")
}

# Install Miniconda (https://docs.conda.io/en/latest/miniconda.html)
if(!require("keras")){
  install.packages("keras")
}

if(!require("tensorflow")){
  install.packages("tensorflow")
  install_tensorflow()
}

use_condaenv("r-tensorflow")

Step 0 set work directories

set.seed(2030)
# setwd("~/Project3-FacialEmotionRecognition/doc")
# here replace it with your own path or manually set it in RStudio to where this rmd file is located. 
# use relative path for reproducibility

Provide directories for training images. Training images and Training fiducial points will be in different subfolders.

Step 1: set up controls for evaluation experiments.

In this chunk, we have a set of controls for the evaluation experiments.

• (T/F) cross-validation on the training set
• (T/F) reweighting the samples for training set
• (number) K, the number of CV folds
• (T/F) process features for training set
• (T/F) run evaluation on an independent test set
• (T/F) process features for test set

Using cross-validation or independent test set evaluation, we compare the performance of models with different specifications. In this part, we tune parameter n.trees and shrinkage for GBM.

Step 2: import data and train-test split

If you choose to extract features from images, such as using Gabor filter, R memory will exhaust all images are read together. The solution is to repeat reading a smaller batch(e.g 100) and process them.

Fiducial points are stored in matlab format. In this step, we read them and store them in a list.

Step 3: construct features and responses

• The follow plots show how pairwise distance between fiducial points can work as feature for facial emotion recognition.

  • In the first column, 78 fiducials points of each emotion are marked in order.
  • In the second column distributions of vertical distance between right pupil(1) and right brow peak(21) are shown in histograms. For example, the distance of an angry face tends to be shorter than that of a surprised face.
  • The third column is the distributions of vertical distances between right mouth corner(50) and the midpoint of the upper lip(52). For example, the distance of an happy face tends to be shorter than that of a sad face.

Figure1

feature.R should be the wrapper for all your feature engineering functions and options. The function feature( ) should have options that correspond to different scenarios for your project and produces an R object that contains features and responses that are required by all the models you are going to evaluate later.

• feature.R
• Input: list of images or fiducial point
• Output: an RData file that contains extracted features and corresponding responses

GBM (Baseline Model)

Step 4: Train a classification model with training features and responses

Call the train model and test model from library.

train.R and test.R should be wrappers for all your model training steps and your classification/prediction steps.

• train.R
  • Input: a data frame containing features and labels and a parameter list.
  • Output:a trained model
• test.R
  • Input: the fitted classification model using training data and processed features from testing images
  • Input: an R object that contains a trained classifier.
  • Output: training model specification
• In this part, we use GBM (baseline model) to do classification.

Model selection with cross-validation

• Do model selection by choosing among different values of training model parameters.

Visualize cross-validation results.

• Choose the “best” parameter value

• Train the model with the entire training set using the selected model (model parameter) via cross-validation.

Step 5: Run test on test images

• evaluation

Summarize Running Time

Prediction performance matters, so does the running times for constructing features and for training the model, especially when the computation resource is limited.

DNN (Dense Neural Network) (Advanced Model)

Step 4: Train a classification model with training features and responses

Call the train model and test model from library.

train.R and test.R should be wrappers for all your model training steps and your classification/prediction steps.

• train.R
  • Input: a data frame containing features and labels and a parameter list.
  • Output:a trained model
• test.R
  • Input: the fitted classification model using training data and processed features from testing images
  • Input: an R object that contains a trained classifier.
  • Output: training model specification
• In this part, we use DNN (advanced model) to do classification.

Model selection with cross-validation

• Do model selection by choosing among different values of training model parameters.

Since DNN training takes very long time in K folds, we do not show the whole tuning process here.

source("../lib/cross_validation_dnn.R")
feature_train = as.matrix(dat_train[, -6007])
label_train = as.integer(dat_train$label)

if(run.cv_dnn){
  # res_cv_dnn <- matrix(0, 1, ncol = 4)
  res_cv_dnn <- matrix(0, nrow = length(lrs), ncol = 5)
  count = 0
  for(i in 1:length(lrs)){
    count = count + 1
    cat("lr =", lrs[i], "\n")
    
    res_cv <- cv.function_dnn(features = feature_train, labels = label_train, K,
                            reweight = sample.reweight, lr = lrs[i])
    
    res_cv_dnn[count,] <- c(lr[i], res_cv[1], res_cv[2], res_cv[3], res_cv[4])
  }
  
  res_cv <- cv.function_dnn(features = feature_train, labels = label_train, K,
                          reweight = sample.reweight, lr = lrs[i])

  colnames(res_cv_dnn) <- c("lr", "mean_error", "sd_error", "mean_AUC", "sd_AUC")
  save(res_cv_dnn, file="../output/res_cv_dnn.RData")
}else{
  load("../output/res_cv_dnn.RData")
}

Visualize cross-validation results.

res_cv_dnn <- as.data.frame(res_cv_dnn)

if(run.cv_dnn){
  p1 <- res_cv_dnn %>%
    ggplot(aes(x = lrs, y = mean_error,
               ymin = mean_error - sd_error, ymax = mean_error + sd_error)) +
    geom_crossbar() +
    theme(axis.text.x = element_text(angle = 90, hjust = 1))
  
  p2 <- res_cv_dnn %>%
    ggplot(aes(x = lrs, y = mean_AUC,
               ymin = mean_AUC - sd_AUC, ymax = mean_AUC + sd_AUC)) +
    geom_crossbar() +
    theme(axis.text.x = element_text(angle = 90, hjust = 1))
  
  print(p1)
  print(p2)
}

• Choose the “best” parameter value

# par_lr_best <- as.numeric(res_cv_dnn[which.min(res_cv_dnn$mean_error), 1])
par_lr_best = 0.0002

• Train the model with the entire training set using the selected model (model parameter) via cross-validation.

source("../lib/train_dnn.R") 
source("../lib/test_dnn.R")

feature_train = as.matrix(dat_train[, -6007])
label_train = as.integer(dat_train$label)
feature_test <- as.matrix(dat_test[, -6007])
label_test <- as.integer(dat_test$label)
label_test <- ifelse(label_test == 2, 1, 0)
weight_test <- rep(NA, length(label_test))

# training weights
weight_train <- rep(NA, length(label_train))
for (v in unique(label_train)){
  weight_train[label_train == v] = 0.5 * length(label_train) / length(label_train[label_train == v])
}

for (v in unique(label_test)){
  weight_test[label_test == v] = 0.5 * length(label_test) / length(label_test[label_test == v])
}

tm_train <- NA
if (sample.reweight){
  tm_train <- system.time(fit_train_dnn <- train_dnn(feature_train, label_train, w = weight_train, feature_test, label_test, weight_test, par_lr_best))
} else {
  tm_train <- system.time(fit_train_dnn <- train_dnn(feature_train, label_train, w = NULL, feature_test, label_test, weight_test = NULL, par_lr_best))
}
save_model_hdf5(fit_train_dnn, "../output/fit_train_dnn.h5")

Step 5: Run test on test images

tm_test = NA
feature_test <- as.matrix(dat_test[, -6007])
if(run.test_dnn){
  fit_train_dnn <- load_model_hdf5("../output/fit_train_dnn.h5")
  tm_test <- system.time({label_pred <- test_dnn(fit_train_dnn, feature_test, type = "predict_classes");})
  prob_pred <- test_dnn(fit_train_dnn, feature_test, type = "predict_proba")
}

• evaluation

## reweight the test data to represent a balanced label distribution
label_test <- as.integer(dat_test$label)

weight_test <- rep(NA, length(label_test))
for (v in unique(label_test)){
  weight_test[label_test == v] = 0.5 * length(label_test) / length(label_test[label_test == v])
}

label_test <- ifelse(label_test == 2, 1, 0)

accu <- sum(weight_test * (label_pred == label_test)) / sum(weight_test)

# prob_pred <- apply(prob_pred, 1, max)
prob_pred <- prob_pred[, 2]

tpr.fpr <- WeightedROC(prob_pred, label_test, weight_test)
auc <- WeightedAUC(tpr.fpr)


cat("The accuracy of model:", "DNN with lr =", par_lr_best, "is", accu*100, "%.\n")
cat("The AUC of model:", "DNN", "is", auc, ".\n")

Summarize Running Time

Prediction performance matters, so does the running times for constructing features and for training the model, especially when the computation resource is limited.

cat("Time for constructing training features=", tm_feature_train[3], "s \n")
cat("Time for constructing testing features=", tm_feature_test[3], "s \n")
cat("Time for training model=", tm_train[3], "s \n") 
cat("Time for testing model=", tm_test[3], "s \n")

###Reference - Du, S., Tao, Y., & Martinez, A. M. (2014). Compound facial expressions of emotion. Proceedings of the National Academy of Sciences, 111(15), E1454-E1462.